Data

Tools

Indexes

Applications

Experiments

Open Source Science

https://github.com/bethgelab/testing_visualizations

Code for the paper " Exemplary Natural Images Explain CNN Activations Better than Feature Visualizations"

https://github.com/bethgelab/testing_visualizations

Science Score: 10.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • ○
    CITATION.cff file
  • ○
    codemeta.json file
  • ○
    .zenodo.json file
  • ○
    DOI references
  • ✓
    Academic publication links
    Links to: arxiv.org
  • ○
    Academic email domains
  • ○
    Institutional organization owner
  • ○
    JOSS paper metadata
  • ○
    Scientific vocabulary similarity
    Low similarity (5.5%) to scientific vocabulary

Keywords

cnn deep-learning interpretiability machine-learning visualizations
Last synced: 5 months ago · JSON representation

Repository

Code for the paper " Exemplary Natural Images Explain CNN Activations Better than Feature Visualizations"

Basic Info
  • Host: GitHub
  • Owner: bethgelab
  • Language: Python
  • Default Branch: master
  • Homepage:
  • Size: 9.02 MB
Statistics
  • Stars: 10
  • Watchers: 10
  • Forks: 2
  • Open Issues: 0
  • Releases: 0
Archived
Topics
cnn deep-learning interpretiability machine-learning visualizations
Created over 5 years ago · Last pushed over 4 years ago
Metadata Files
Readme

README.md

Code for: Exemplary Natural Images Explain CNN Activations Better than Feature Visualizations

This information contains code to reproduce the experiments described in the paper Exemplary Natural Images Explain CNN Activations Better than Feature Visualizations by Judy Borowski, Roland S. Zimmermann, Judith Schepers, Robert Geirhos, Thomas S. A. Wallis, Matthias Bethge and Wieland Brendel. If there are any questions, please open an issue here on GitHub, and we'll try to answer it.

Code to perform the stimuli selection and creation can be found in the folder generate_stimuli. Please see this folder for more information on how to use the code.

The code to conduct the actual psychophysical experiment will be added soon.

Citation

@misc{borowski2020exemplary,
    title={Exemplary Natural Images Explain CNN Activations Better than Feature Visualizations}, 
    author={Judy Borowski and Roland S. Zimmermann and Judith Schepers and Robert Geirhos and Thomas S. A. Wallis and Matthias Bethge and Wieland Brendel},
    year={2020},
    eprint={2010.12606},
    archivePrefix={arXiv},
    primaryClass={cs.CV}
}

Owner

  • Name: Bethge Lab
  • Login: bethgelab
  • Kind: organization
  • Location: Tübingen

Perceiving Neural Networks

GitHub Events

Total
Last Year

Dependencies

human_experiments/requirements.txt pypi
  • GitPython *
  • Pillow *
  • PsychoPy ==2020.1.2
  • PyOpenGL ==3.1.5
  • PyQt5 ==5.12.3
  • PyQt5-sip ==4.19.18
  • PyQtChart ==5.12
  • PyQtWebEngine ==5.12.1
  • PySocks ==1.7.1
  • PyYAML ==5.4
  • Pygments ==2.7.4
  • Pypubsub ==4.0.3
  • SoundFile ==0.10.2
  • arabic-reshaper ==2.0.15
  • astroid *
  • astunparse ==1.6.2
  • backcall ==0.2.0
  • brotlipy ==0.7.0
  • certifi ==2020.6.20
  • cffi ==1.14.0
  • chardet ==3.0.4
  • configobj ==5.0.6
  • configparser ==5.0.0
  • cryptography ==3.3.2
  • cycler ==0.10.0
  • decorator ==4.4.2
  • distro *
  • dukpy ==0.2.2
  • esprima ==4.0.1
  • et-xmlfile ==1.0.1
  • freetype-py ==2.1.0.post1
  • future ==0.18.2
  • gevent *
  • gitdb *
  • glfw *
  • glob2 ==0.7
  • greenlet *
  • idna *
  • imageio *
  • imageio-ffmpeg *
  • ipython *
  • ipython-genutils ==0.2.0
  • isort ==4.3.21
  • javascripthon ==0.11
  • jdcal ==1.4.1
  • jedi *
  • json-tricks ==3.15.2
  • kiwisolver ==1.2.0
  • lazy-object-proxy ==1.4.3
  • lxml ==4.6.3
  • macropy3 ==1.1.0b2
  • matplotlib *
  • mccabe ==0.6.1
  • mock *
  • moviepy ==1.0.1
  • msgpack ==1.0.0
  • msgpack-numpy *
  • msgpack-python ==0.5.6
  • numexpr ==2.7.1
  • numpy *
  • olefile ==0.46
  • openpyxl *
  • pandas *
  • parso ==0.7.0
  • pathlib2 ==2.3.5
  • pexpect ==4.8.0
  • pickleshare ==0.7.5
  • proglog ==0.1.9
  • prompt-toolkit ==3.0.5
  • psutil ==5.7.0
  • psychtoolbox ==3.0.16
  • ptyprocess ==0.6.0
  • pyOpenSSL ==19.1.0
  • pycparser *
  • pygame ==1.9.6
  • pyglet *
  • pylint *
  • pyo ==1.0.1
  • pyosf ==1.0.5
  • pyparallel ==0.2.2
  • pyparsing ==2.4.7
  • pyserial ==3.4
  • python-bidi ==0.4.2
  • python-dateutil ==2.8.1
  • python-gitlab *
  • python-xlib ==0.27
  • pytz ==2020.1
  • pyzmq ==19.0.1
  • questplus ==2019.4
  • requests *
  • rope ==0.17.0
  • scipy *
  • six *
  • smmap *
  • sounddevice ==0.3.15
  • tables ==3.6.1
  • toml *
  • tornado ==6.0.4
  • tqdm *
  • traitlets ==4.3.3
  • typed-ast ==1.4.1
  • typing-extensions ==3.7.4.2
  • urllib3 ==1.25.9
  • wcwidth *
  • webassets ==2.0
  • wrapt ==1.11.2
  • wxPython ==4.0.7.post2
  • xarray ==0.15.1
  • xlrd ==1.2.0
  • zope.event *
  • zope.interface ==5.1.0